The task at hand was to seamlessly integrate the open-source ChatGPT functionality into a personalized React app, paving the way for the creation of OKR Copilot, a revolutionary tool designed to revolutionize the way our client sets and achieves their objectives. This is how it went.

Client and goal

The client was the Croatian business consulting company Best Advisory. They are a company that helps small and medium businesses become more successful by guiding them to achieve digital transformation, offers strategic consulting, and helps achieve sustainable and responsible business. 

They already have numerous partners and clients. Best Advisory approached us with a request to build AI-powered software that will help their clients create reasonable and achievable key performance indicators. 

This idea was born from a simple need – you don’t need years of experience to create achievable OKRs (Objectives and Key Results), but you can use access to AI to create some for you. That is how the idea of OKR-copilot moved into the phase of a project.

Product Description

OKR Copilot is a React web app, which uses the OpenAI API to generate precise OKR-s. It allows users to be as precise as possible by selecting their business vertical, their previous KPIs, their business goals, the number of employees, etc. By imputing this information, the user gets a precise response from the AI service with suggestions of what OKRs should the business focus on in the next period. 

Applied technologies

Here are the technologies our team applied to deliver the app 

React – React is used for building user interfaces, especially for single-page applications where components can be reused to manage and update the UI efficiently. The frontend of OKR Copilot application is written using React as a programming language.

Java – Java is a versatile, object-oriented programming language used for building a wide range of applications. It is known for its portability and platform independence. We used Java to build the backend of OKR Copilot application.

Firebase – Firebase is a platform developed by Google that provides various services for developing web and mobile applications. It includes features like authentication, real-time database, cloud functions, hosting, and more. In order to keep the application secure, we used Firebase authentication for user management. 

OpenAI –  It provides open-source tools and APIs for building and experimenting with AI models. We used Open AI capabilities to prompt responses for creating the OKR-s. 

Typescript – TypeScript is a statically typed language that compiles to plain JavaScript. It adds static typing and other features to help developers catch errors during development. Typescript was used for ease of programming.

SQL – SQL is a domain-specific language used for managing and manipulating relational databases. It is used to define, query, and manipulate data in databases. This was our primary tool for constructing the database which will contain all the information relevant to the users.

OKR copilot login page

Development process

The project idea was simple enough, with enough challenges to bring the unavoidable complexity. 

Firstly, to be able to do anything, we needed to allow users to register and log into the app. This was done with Google’s Firebase service for authenticating users. We also allowed the users to sign up for the app with their Google account. This was the standard procedure our company uses in most of the projects we do. 

Secondly, we needed to connect the OpenAI with our system, via their open API. 

The most important process that needed to be created was the prompt for the ChatGPT. The response from the AI needed to be personalized for each user, depending on multiple factors that are unique for each company. These factors include the size of the company, location, and department for which the OKRS are being generated. Also, we needed to guide the AI to generate OKRs that will be compatible with the company’s vision, as well as their industry vertical and strategic goals. 

Thirdly, in cooperation with the client, we decided that it would be a good idea to generate three responses so that users could choose between the generated responses. Afterward, this feature was upgraded to allow users to generate up to 10 responses if they were not satisfied with the results in the first place. 

Lastly, we added the feature to send the responses via email.

OKR Copilot

Next steps

Once we achieved functional requirements, it was time to make the app more visually appealing to the clients. With the help of a design team, the app was transformed from the standard, plain industrial software to a more modernized version, with stunning graphical elements. 

We also added a landing page. During one of our meetings, we found out that the app will not be a self-explanatory tool for new clients that might come without first corresponding with Best Advisory. We developed a new and modern landing page that connects the app and the Best Advisory web page. 

OKR Copilot homepage

How does software such as OKR Copilot improve the company’s business?

On the side of the end user, this software allows them to create OKRs according to their own company needs and strategic goals. It is quick as a wink: after filling out a short, 5-minute questionnaire, you get OKR-s generated for you. This reduces the time needed for establishing business objectives.

On the other hand, it provides a chance for new companies to connect and gain reliable partners in their field of work. 

The question that naturally arises from this is ‘Why would anyone use the software instead of prompting the chat GPT by himself?’ The simplest answer is that it would take much more time to prompt the OpenAI to respond with quality than it would take to write them on paper! By this, we mean that the prompt in OKR Copilot is programmed to return quality responses with formatting which allows it to save time. 

Recommended ➜ Read about our WooCommerce delivery plugin

How much time does it take to build this type of software?

This particular project was done in around three weeks of development time. This includes the visual redesign of the app. 

Developed and redesigned in just three weeks, this React web app, powered by OpenAI, isn’t just a tool – it offers an insight into how businesses will be managed in the future.

As the digital space becomes increasingly focused on efficiency and practicality, tools like OKR Copilot reflect a broader shift towards user-friendly, agile solutions. There’s a growing demand for instant tools that streamline complex processes. Through this project, we learned that it can be done fast, reliably, and efficiently. 

One can predict that the future will bring apps that seamlessly integrate into daily operations. Two things are important. Speedy development, but also the quality of useful intelligence within these tools. 

The ability to have precise OKRs with just a few clicks speaks to the changing ways of doing business, with user-friendly, AI-driven applications.

As you explore the possibilities of B2B digital transformation, keep in mind that it sets you apart from your competition, but more so it helps your business breathe and thrive. 

Are there any quick solutions (or intricate ones) that would help you streamline your processes today? 

Connect with us for a free consultation, Lumen Spei is your B2B digital transformation partner. 


Miloš Videnović

Project manager at Lumen Spei⎜Passionate about translating project visions and objectives into reality, offering tangible results in line with client expectations.